How do you know if the disks you will be using from ANY particular vendor can muster up the IOPS and MBPS required to satisfy your current or future workloads?

Some simple calculations can help you in this regard.

We have talked about these a little in a previous post but I thought it necessary to have a single post that had the definitions and calculations all in one place:

IOPS-I/O Per Second-the number of I/O operations that can perform in one second.
MBPS-Mega Bytes Per Second-throughput and equates to the Bytes Per Second a storage device can stream.
RPM-Revolutions Per Minute.-The number of rotations completed in one minute.
Seek Time-The time required to position the head over the target track. (less for read than write)
Average Latency-(rotational Latency) Average time after head is over the track for the target sector to rotate under the head before a read or write.

Full Rotation	(1/ [RPM/60000] )
Rotational Latency	(Full Rotation)/2
Average Seek Time	(4.2+4.7)/2
IO Time	(Rotational Latency + Average Seek Time)
IOPS	(1/IO Time)*1000

Number of Disks	obviously 1 if you care about just one disk, but use a number that will be involved in all I/O
Segment Size	what size I/O you will be using for large reads, configurable on your SAN, application, or O/S
MBPS	(IOPS * Number of Disks * Segment Size) / 1024

That's all there is to it! Quite easy and you should bookmark this page or print off somewhere for reference. Now go off, do the calculations for your particular disk subsystem and post the results here. Sounds like fun!

If you have an Oracle database, did you run the scripts given a couple of days ago from this blog? I would also be very curious how those numbers match up with the numbers above. You should be at least.